Triceps and subscapular skinfold thickness percentiles and cut-offs for overweight and obesity in a population-based sample of schoolchildren and adolescents in Bogota, Colombia Academic Article


  • The assessment of skinfold thickness is an objective measure of adiposity. The aims of this study were to establish Colombian smoothed centile charts and LMS L (Box–Cox transformation), M (median), and S (coefficient of variation) tables for triceps, subscapular, and triceps + subscapular skinfolds; appropriate cut-offs were selected using receiver operating characteristic (ROC) analysis based on a population-based sample of children and adolescents in Bogotá, Colombia. A cross-sectional study was conducted in 9618 children and adolescents (55.7% girls; age range of 9–17.9 years). Triceps and subscapular skinfold measurements were obtained using standardized methods. We calculated the triceps + subscapular skinfold (T + SS) sum. Smoothed percentile curves for triceps and subscapular skinfold thickness were derived using the LMS method. ROC curve analyses were used to evaluate the optimal cut-off point of skinfold thickness for overweight and obesity, based on the International Obesity Task Force definitions. Subscapular and triceps skinfolds and T + SS were significantly higher in girls than in boys (p < 0.001). The ROC analysis showed that subscapular and triceps skinfolds and T + SS have a high discriminatory power in the identification of overweight and obesity in the sample population in this study. Our results provide sex- and age-specific normative reference standards for skinfold thickness values from a population from Bogotá, Colombia.

publication date

  • 2016/1/1


  • 8


  • Adiposity
  • Advisory Committees
  • Colombia
  • Cross-Sectional Studies
  • Obesity
  • Population
  • ROC Curve
  • Skinfold Thickness
  • adiposity
  • cross-sectional studies
  • gender
  • methodology
  • obesity
  • reference standards
  • sampling
  • school children
  • skinfold thickness

International Standard Serial Number (ISSN)

  • 2072-6643

number of pages

  • 10

start page

  • 1

end page

  • 10